This book provides a comprehensive overview of statistical methods for analyzing discrete longitudinal data, with a focus on non-Gaussian models. It builds upon the linear mixed model, which is widely used for continuous longitudinal data, and extends it to handle binary, ordinal, and count data. The authors, Geert Molenberghs and Geert Verbeke, emphasize practical applications over theoretical rigor, offering a wide range of examples and case studies. The text covers various model families, including marginal, conditional, and subject-specific models, and discusses their strengths and limitations. It also addresses issues such as handling missing data, inference for variance components, and the interpretation of model results. The book includes detailed explanations of statistical software tools like SAS, with specific examples of how to implement these models. It also explores advanced topics such as generalized estimating equations (GEE), pseudo-likelihood methods, and non-linear mixed models. The authors highlight the importance of choosing the appropriate model based on the data structure and research question, and provide guidance on model selection and interpretation. The book is intended for applied statisticians, biomedical researchers, and professionals in the biopharmaceutical industry, as well as those involved in medical and public health research. It includes a variety of real-world examples and case studies, and is structured to facilitate both learning and application of the methods discussed. The authors also acknowledge the contributions of many colleagues and students who have helped shape the content of the book.This book provides a comprehensive overview of statistical methods for analyzing discrete longitudinal data, with a focus on non-Gaussian models. It builds upon the linear mixed model, which is widely used for continuous longitudinal data, and extends it to handle binary, ordinal, and count data. The authors, Geert Molenberghs and Geert Verbeke, emphasize practical applications over theoretical rigor, offering a wide range of examples and case studies. The text covers various model families, including marginal, conditional, and subject-specific models, and discusses their strengths and limitations. It also addresses issues such as handling missing data, inference for variance components, and the interpretation of model results. The book includes detailed explanations of statistical software tools like SAS, with specific examples of how to implement these models. It also explores advanced topics such as generalized estimating equations (GEE), pseudo-likelihood methods, and non-linear mixed models. The authors highlight the importance of choosing the appropriate model based on the data structure and research question, and provide guidance on model selection and interpretation. The book is intended for applied statisticians, biomedical researchers, and professionals in the biopharmaceutical industry, as well as those involved in medical and public health research. It includes a variety of real-world examples and case studies, and is structured to facilitate both learning and application of the methods discussed. The authors also acknowledge the contributions of many colleagues and students who have helped shape the content of the book.